Predicting Tumor Metastasis by Employing a Target Organ/Primary Lesion Fusion Radiomics Model

NCT ID: NCT06202404

Last Updated: 2024-02-09

Basic Information

• Recruitment Status: RECRUITING
• Total Enrollment: 166 participants
• Study Classification: OBSERVATIONAL
• Study Start Date: 2024-01-01
• Study Completion Date: 2027-12-31

Brief Summary

A pre-metastatic target organ/primary lesion fusion radiomics model was developed based on the "soil-seed" theory to predict comman tumor metastasis in retrospective settings. To prospectively verify the performance of the target organ/primary lesion fusion radiomics model in predicting tumor metastasis patterns (brain metastasis in lung cancer, liver metastasis in colorectal cancer, lung metastasis in breast cancer), we designed this prospective observational trial.

Detailed Description

Metastasis is an important biological feature of malignant tumors, and it is also the main factor affecting the outcome of cancer patients. The direct death of most cancer patients is the mass effect of metastatic lesions or the loss of metastatic target organ function. Therefore, it has become one of the urgent and arduous tasks in the field of oncology to explore the pathophysiological mechanism of malignant tumor metastasis and find new targets to form new methods for the early diagnosis and treatment of tumor metastasis.

Tumor metastasis is a continuous and gradual process. Tumor cells in the primary tumor proliferated, tumor cells shed from the primary tumor, invaded the basement membrane, invaded blood vessels, lymphatic vessels or body cavity, and migrated to another site or organ with the blood flow and lymph flow. Tumor cells adhered to the capillaries of the target organ, penetrated the blood vessels to form micrometastases, cell proliferation and produced new blood vessels, and formed secondary tumors of the same type as the primary tumor. They can invade and metastasize again. In the process of metastasis, gene mutations and subsequent changes in the immunogenicity of tumor cells promote the formation of local tumor tissue inhibitory microenvironment, which suppresses the host immune system and leads to tumor metastasis.

Previous studies have found that tumor metastasis has Organ specificity, called "Organ-tropism". That is, different types of tumors tend to metastasize to different organs. For example, lung cancer often metastasizes to the brain, colorectal cancer and pancreatic cancer commonly metastasize to the liver, breast cancer and prostate cancer commonly metastasize to the bone, and so on. Therefore, Stephen Paget, a British pathologist, put forward and elaborated his "seed and soil theory" more than a hundred years ago, advocating that metastasis requires the spread of tumor cells, that is, "seeds", and the ideal environment for receiving organs, that is, "fertile soil". He proposed in this theory that successful metastasis to distant organs requires not only tumor cells with metastatic potential (seeds) but also a suitable host organ (soil). During tumor progression, the composition and phenotypic types of immune cells in the tumor microenvironment are shaped by both malignant cells and the niche of adjacent tissues. Metastatic tumors (seeds) will retain genetic characteristics when they are transferred to different organ-specific tissue environments (soil). It is not enough for tumor cells to reach a certain organ tissue, but also requires the coordination between tumor cells (seeds) and organ microenvironment (soil) to plant and grow in a suitable environment (soil) and become metastatic foci. In the "seed and soil theory", the intrinsic characteristics of seeds have been proved to affect various key transfer processes. On the other hand, few studies have focused on the relationship between the intrinsic characteristics of soil (target organ) and the formation of malignant tumor metastasis, one of the main reasons is that it is often not ethical to conduct invasive research on the target organ of patients without metastasis.

Radiomics is an emerging field, which develops new biomarkers through quantitative analysis of imaging examination images of patients, which can study and analyze the potential pathophysiological characteristics of regions of interest in a non-invasive way. Previous studies have shown that radiomics studies of seeds (primary tumors) can predict metastasis of many malignant tumors, which establishes radiomics evidence for the "seed and soil" theory from the perspective of seeds. However, radiomics studies from the perspective of soil (the target organ for metastasis) are rare. Taking the brain metastases of lung cancer as an example, most existing radiomics studies using brain imaging information only evaluate the regions where brain metastases have occurred. Considering that radiomics can also extract biological features from target organs that have not yet metastasized, our group has found that radiomics analysis of pretreatment brain MRI images of lung cancer patients without baseline brain metastasis can effectively predict the probability of brain metastasis in a certain period of time. This study also revealed for the first time that the target organ without metastasis (soil) also plays a key role in the occurrence of malignant tumor metastasis.

The results of this study also suggest that the malignancy prediction model derived from radiomics could be a promising tool for screening patients at high risk of tumor metastasis. Integrated risk prediction models that take into account seed and soil characteristics may help guide clinical practice (e.g., more rigorous monitoring of high-risk patients or the use of more intensive systemic treatment).

In conclusion, it is necessary to further explore the applicability of this seed and soil theory-based radiomics model for cancer prediction to verify its universality and clinical application prospects. In this study, we will collect imaging information of patients' baseline primary lesions and potential metastatic target organs in a variety of common tumors (lung cancer, colorectal cancer, breast cancer, etc.). To explore the performance of the "seed and soil" radiomics model in predicting cancer metastasis patterns (brain metastasis of lung cancer, liver metastasis of colorectal cancer, lung metastasis of breast cancer), in order to establish a non-invasive, cost-effective and effective radiomics prediction model for distinguishing "high risk" and "low risk" cancer metastasis in corresponding cancer types. To provide a theoretical basis for the further formation of individualized diagnosis and treatment standards for patients with malignant tumors.

Conditions

Metastasis; Predictive Cancer Model

Study Design

• Observational Model Type: COHORT
• Study Time Perspective: PROSPECTIVE

Study Groups

High risk group

This is a observational study, patients will be assigned to high-risk or low-risk groups (according to a radiomics model) by the investigators. Both the patients and the treating physicians will be blinded to this procedure. No interventions excluding standard treatment procedure executed by the treating physicians.

No interventions assigned to this group

Low risk group

This is a observational study, patients will be assigned to high-risk or low-risk groups (according to a radiomics model) by the investigators. Both the patients and the treating physicians will be blinded to this procedure. No interventions excluding standard treatment procedure executed by the treating physicians.

No interventions assigned to this group

Eligibility Criteria

Inclusion Criteria

1. ECOG performance status score 0-2;

2. Histologically or cytologically confirmed stage III-IV NSCLC;

3. If the baseline pathology is adenocarcinoma, driver gene testing (at least EGFR/ALK/ROS1/KRAS/MET) should be performed;

4. Complete imaging data of baseline stage (contrast-enhanced MR For lung cancer, contrast-enhanced MR/CT for colorectal cancer, and chest CT for breast cancer);

5. No target organ metastasis on baseline imaging (no brain/liver/lung metastasis for lung cancer/colorectal cancer/breast cancer, respectively);

6. Patients received at least one systemic therapy (chemotherapy, targeted therapy, immunotherapy, etc.) and received regular follow-up;

7. Regular follow-up during and after treatment;

8. Life expectancy ≥6 months;

Exclusion Criteria

1. Patients with indeterminate pathological type;

2. Patients without baseline imaging data before treatment;

3. Baseline imaging examination showed that the corresponding target organ had metastasis (lung cancer/colorectal cancer/breast cancer corresponding to brain/liver/lung metastasis);

4. patients who cannot or refuse to receive regular imaging follow-up;

5. Combined history of other malignant tumors;

6. Medical examination or clinical findings or other uncontrollable conditions that the investigator considers may interfere with the results or increase the risk of treatment complications for the patient; ,

7. Lactating or pregnant women;

8. Receiving other long-term medications that may affect disease progression as assessed by a physician.

• Minimum Eligible Age: 18 Years
• Maximum Eligible Age: 75 Years
• Eligible Sex: ALL
• Accepts Healthy Volunteers: No

Sponsors

Fudan University

OTHER

Sponsor Role: lead

Responsible Party

Zhengfei Zhu

Professor

Responsibility Role: PRINCIPAL_INVESTIGATOR

Locations

Fudan University Shanghai Cancer Center

Shanghai, Shanghai Municipality, China

Site Status: RECRUITING

Fudan University Shanghai Cancer Center

Shanghai, , China

Site Status: RECRUITING

Countries

China

Central Contacts

Xiao Chu

Role: CONTACT

chuxiao@sibs.ac.cn

15821383376

Facility Contacts

Zhengfei Zhu, MD

Role: primary

fuscczzf@163.com

+86 18017312901

Xiao Chu

Role: backup

chuxiao@sibs.ac.cn

15821383376

Zhengfei Zhu, MD

Role: primary

fuscczzf@163.com

+86-18017312901

References

Chu X, Gong J, Yang X, Ni J, Gu Y, Zhu Z. A "Seed-and-Soil" Radiomics Model Predicts Brain Metastasis Development in Lung Cancer: Implications for Risk-Stratified Prophylactic Cranial Irradiation. Cancers (Basel). 2023 Jan 2;15(1):307. doi: 10.3390/cancers15010307.

Reference Type: BACKGROUND

PMID: 36612303 (View on PubMed)

Other Identifiers

2023-Seedandsoil01

Identifier Type: -

Identifier Source: org_study_id